Lead Session Chair:
Stephan Barth, Managing Director, ForWind - Center for Wind Energy Research, Germany
Estivaliz Lozano-Minguez (1) F P Feargal Brennan (1) Athanasios Kolios (1)
(1) Cranfield University, Cranfield, United Kingdom
Printer friendly version: Print
Download poster(0.84 MB)
Presenter's biographyBiographies are supplied directly by presenters at OFFSHORE 2015 and are published here unedited
Estivaliz obtained a double degree in Civil Engineering from the Universidad Politécnica de Valencia, and an MSc in Economics for Natural Resource and Environmental Management from Cranfield University. MSc thesis entitled “A multi-criteria analysis of the optimal offshore wind turbine support structures”. In 2011, she started a PhD in the Offshore Renewable Energy Group entitled “Reliability based structural optimization for a new generation of marine and offshore structures”. This research intends to develop a generic design process able to define the optimum structure subjected to loads of highly stochastic nature, and will mainly focus on the design of complex structures.
Fitting new S-N curves for welded tubular joints.
Energy forecasts predict significant growth in demand in the coming decades, caused by increasing economic growth rates all over the world but especially in developing countries. Fossil fuels – coal, oil, and gas – supplied around 80 percent of world primary energy demand in 2010. It is an essential requirement to make the renewable energy industry more economically efficient in order to greater investment in the sector and to reverse those percentages.
Extending the service life of the offshore wind turbine support structures would contribute to the reduction of the costs. The S-N curve approach is used in the fatigue-life assessment at the design stage, in particular the T' curve is applied for the fatigue behaviour of welded tubular joints. However, T' curve does not make differentiation between joint types or applied loads, which affects the fatigue strength leading to obtain a value which may not be the optimum for predicting the fatigue life.
Main body of abstract
This paper analyses the effectiveness of the T' curve and attempts to obtain the curve which best sets the available data. In the regression analysis for obtaining the T' curve, 59 samples of 16 mm thick tubular joints in were used but only 44 of these samples were T-joints. The statistical regression package IBM SPSS was used for the statistical assessment. The Least Squares Method (LSM) and the Maximum Likelihood Method (MLM) were applied for estimating the regression coefficients.
The MLM curve seems to be the best fit curve for the available data. This study should not be used to assess fatigue, a more comprehensive study should be undertaken.
The results should be used to show that it is needed to continue working on the development of new curves to reduce safety factors and thus reduce costs, without incurring in an increase of the probability to failure.